Nearly one in every eight children is born prematurely in the US, yet very little is known about visual development or patterns of vision damage in this large group of children. The leading cause of bilateral low vision in children in the US is cerebral visual impairment (CVI), usually caused by neurological injury in premature infants. This condition will be potentially preventable or amenable to treatment in the near future as new therapies emerge for the prevention and treatment of neurological disease that result from extreme prematurity. Clearly, better methods to test and assay vision are needed. Otherwise, there will be no accurate way to judge the effects of medical or rehabilitative intervention. In this study, we will measure visual development in healthy, premature infants less than 32 weeks gestational age, and compare this to full term infants. Sweep VEP grating acuity, vernier acuity, and contrast sensitivity measures will be obtained at 6, 9, and 12 months chronological age in the premature group, and at 3, 6, 9, and 12 months chronological age in the fullterm group. In a second aim, we will measure these same three visual functions at 6, 9, and 12 months in infants who suffer an intraventricular hemorrhage (IVH) as a complication of prematurity. Results from this study will provide important information about visual development in normal premature infants, and the manner in which such visual development is perturbed by extreme prematurity and neurological damage (IVH). In a third aim, we will evaluate the role of diffusion tensor imaging (DTI) in identifying infants with vision abnormalities. DTI is performed early in the infant's life, at a time when intervention is most likely to prevent or treat neurological damage. By performing a structure-function analysis, comparing the spectrum of acuity changes in children with the spectrum of changes on their scans, we expect to be able to identify those children who have suffered damage to their visual system. This will allow early detection of vision abnormalities and pave the way for early medical and rehabilitation efforts. A comparison of head ultrasound and DTI findings will demonstrate whether DTI should replace ultrasound for the diagnosis of neurological injury in the newborn nursery. [unreadable] [unreadable]